© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 45797
    Prev
    Next
    INDEX
    Explanations

    phrases related to describing the characteristics or attributes of various subjects

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.15
    0.6%
    188
    +0.14
    0.6%
    575
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.15
    0.17
    608
    +0.14
    0.18
    393
    +0.13
    0.20
    Negative Logits
    enger
    -0.84
    rom
    -0.83
    ogun
    -0.83
    oning
    -0.81
    oned
    -0.81
    itone
    -0.79
    visors
    -0.78
    ron
    -0.75
    inis
    -0.75
    visor
    -0.75
    POSITIVE LOGITS
    tenance
    0.74
     incarn
    0.72
     occurrences
    0.70
    falls
    0.70
    entially
    0.68
     intertwined
    0.68
     nature
    0.68
    CLASSIFIED
    0.67
     insepar
    0.67
     preserves
    0.66
    Activations Density 6.462%

    No Known Activations